Balancing Efficiency and Environmental Impacts in Greek Viticultural Management Systems: An Integrated Life Cycle and Data Envelopment Approach

Abstract

Greek wines excel in quality and exports, but the viticultural sector faces significant challenges from complex supply chains, shifting European policies, and the growing need for sustainability amidst climate change and economic pressures. External environmental costs could affect significantly the decision-making process of farmers, reflecting a broader evaluation of sustainability in viticulture. This study evaluates the economic and environmental impacts of organic, integrated, and conventional viticulture management systems in Drama, Greece using a life cycle (LC) approach and data envelopment analysis (DEA) to determine efficiency, quantify environmental impacts in monetary terms, and incorporate these costs into the analysis. Organic management systems have lower energy consumption and emissions compared to integrated and conventional systems, with organic systems ranging from 4546 to 6573 kWh/ha in energy use and 1358 to 1795 kg CO₂ eq./ha in emissions, while integrated and conventional systems range from 9157 to 12,109 kWh/ha and 2961 to 3661 kg CO₂ eq./ha. The DEA analysis reveals that most organic systems perform efficiently when accounting for environmental costs, whereas conventional systems face significant efficiency declines, with only a few maintaining optimal performance. Policy-supported transitions based on the provider gets principle are crucial for balancing economic and environmental goals in viticulture, as the integration of shadow prices significantly impacts efficiency.

1. Introduction

Viticulture significantly contributes to various sectors of the Greek economy, offering high-quality grapes and wines while also enhancing tourism through attractions like the Wine Roads [1]. The Greek wine industry is characterized by a mix of large, medium, and small wineries, along with associations of agricultural cooperatives, while the primary export destinations (Germany and France) account for 49.7% and 13.2% of Greek wine exports, respectively [2]. Furthermore, Greek wines are recognized for their high quality, with future projections indicating that increasing growing season temperatures and mild water deficits could enhance the potential for producing superior vintages [3]. However, the Greek wine supply chain is highly complex, giving rise to a range of structural and situational challenges shaped by various socio-economic and environmental factors [4,5]. These challenges are intricately linked to the diverse stakeholders involved, the variability in production practices, and the impact of external market conditions.

The Common Agricultural Policy (CAP) 2023-27 focuses primarily on subsidiary measures integrating heightened environmental objectives [6] and neglecting concerns of farmers over rising production costs, unfair competition from imports, and undervalued EU-grown food products [7]. Viticulture is significantly impacted by these issues, being one of the sectors most influenced by evolving governmental policies throughout history [8]. In response to the challenges posed by climate change and the need to preserve natural resources, a range of viticultural management strategies has been implemented (e.g., organic viticulture). These approaches aim to improve species and ecosystem biodiversity, while addressing the increasing demand for organically grown grapes [9,10,11]. Nevertheless, these viticultural systems must be carefully designed to ensure the provision of adequate economic profitability for farmers, while optimizing input usage. This is essential for achieving social acceptability and environmental sustainability.

Organic, integrated, and conventional farming systems impact environmental sustainability based on regional and site-specific pedo-climatic factors [12]. Organic systems generally result in lower yields due to a higher prevalence of diseases compared to integrated [13] and conventional systems [14], but offer superior environmental performance [15,16,17,18]. In similar fashion, conventionally farmed vineyards typically emit higher levels of pollutants, primarily due to the extensive use of pesticides and fertilizers [19,20]. Nevertheless, other studies demonstrate significant variability in yield between organic and conventional treatments, largely influenced by the specific site characteristics of each vineyard [20,21], thereby creating a challenging environment for comparing different viticultural farming systems. Another point of contention is the grape juice, with organic grapes showing greater resistance to deterioration compared to those from integrated vineyards [22].

In this context, sustainable management of farming systems necessitates a comprehensive and interdisciplinary approach to optimize the efficient use of water, energy resources, and inputs [23]. In order to measure efficiency of agriculture, an assortment of measurement methods and indicators have been highlighted by the Food and Agriculture Organization (FAO) [24]. Data envelopment analysis (DEA) is widely employed in agriculture to assess the efficiency of agricultural systems [25,26], given its capability to handle multiple inputs and outputs, which makes it especially well-suited for this sector [27].

Nevertheless, the viticultural sector is experiencing a crucial maturation phase, grappling with environmental challenges and life cycle assessment (LCA) provides a comprehensive approach to evaluating environmental impacts. To effectively evaluate multiple input/output data across a large number of similar entities, the combined use of LCA and DEA has become widely adopted in agricultural studies. This integrated approach serves as a valuable tool for comprehensively assessing both environmental and operational performance across numerous comparable entities [28].

However, quantifying the environmental impacts of agricultural practices and translating these impacts into economic terms is challenging. Policymakers, farmers, and consumers are increasingly evaluating a wide range of economic, external, and environmental factors rather than focusing solely on quality and price [29,30,31]. Environmental prices represent the constructed monetary value assigned to the social cost of pollution, by quantifying the economic welfare loss associated, typically expressed in Euros per kilogram of pollutant [32]. Another approach to convert environmental impacts into monetary terms is through the implementation of shadow prices, integrating environmental considerations into economic analyses and investment decision-making [33]. Thus, the core of shadow prices lies in calculating the actual costs by incorporating external costs, thereby providing additional insights for decision-making.

The current study aims to evaluate the economic and environmental impacts of three different viticulture management systems—organic, integrated, and conventional—in Drama, Greece using a life cycle (LC) approach. Additionally, the efficiency of each vineyard production plan will be assessed through DEA to identify the most efficient management systems. In this study, environmental impacts are also quantified in monetary terms by converting CO₂ equivalents into costs, which are then incorporated into the results. This allows for an investigation of whether efficiency remains consistent or changes when shadow prices are factored into the decision-making process. Notably, this study is the first to integrate LCA and DEA across various viticultural systems, emphasizing the efficiency of each system by including social costs in the form of shadow prices.

2. Materials and Methods

2.1. Case Study Area and Description of Management Systems

The region of Drama is known for cultivating indigenous Greek grape varieties, such as Assyrtiko, Malagousia, Vidiano, Agiorgitiko, Xinomavro, Moschomavro, and Mavrotragano. These varieties produce wines with distinct flavours that reflect the region’s terroir. However, the region also excels in growing international grape varieties like Chardonnay, Sauvignon Blanc, Cabernet Sauvignon, Merlot, and Syrah. These varieties are often blended with local grapes to create unique wines with both Greek and international appeal. The vineyards assessed herein are situated in Northern Greece (Figure 1), regional unit of Drama which produces a plethora of PGI-labelled wines stemming from several wineries and vineyards, operating in the wider area [34]. The region is renowned for its rich history in viticulture dating back to ancient times. However, over the last 50 years, the region has emerged as a modern wine-producing area, with significant investments in vineyards and wineries which resulted in a rapidly growing reputation for producing high-quality wines [35].

In the current study, wine-producing vineyards were assessed from each of the three underlying management systems, organic (O), integrated (I), and conventional (C) vineyard management. Organic viticulture is a method of grape growing that implements sustainable, environmentally friendly practices without the use of synthetic chemicals. Its core principles include maintenance of soil health and the use of biological controls and organic-approved substances to manage pests and diseases instead of synthetic pesticides, to safeguard biodiversity and environmental sustainability [36]. Organic viticulture employs practices that conserve water, such as drip irrigation and adheres to organic certification standards [37]. Conventional management, on the other hand, entails the intensive use of all available resources, like water and chemical fertilizer inputs, crop mechanization, genetic modification, and chemical (or natural as well) pesticide products to maximize crop yield in terms of productivity and efficiency per unit of land [38]. Integrated management lies somewhere in between organic and conventional management. Integrated management concerns a holistic approach to vineyard management that combines the best practices of conventional, organic, and sustainable farming to achieve optimal grape production while minimizing environmental impact [39]. It entails integrated pest management (IPM), integrated nutrient management (INM), and precision agriculture technologies to achieve an acceptable crop yield in terms of quantity and quality with the lowest possible chemical inputs aiming at economic and environmental balance [40,41].

2.2. System Boundaries

The primary objective of this study is to conduct a comprehensive environmental and economic evaluation of three vineyard management systems: organic, integrated and conventional. Additionally, the study compares the efficiency of the management systems during the full production stage by developing case-specific models to the Prefectural Unit of Drama, Greece. Each management system utilizes varying input levels for different operations, including fertilization, harvesting, irrigation, pruning, agrochemical application, tying, topping, weed control, and defoliation. A three-year life cycle during the full production stage, using a cradle-to-factory gate approach, was selected as the basis for data collection. This focus on the on-field activities aims to avoid biases related to the impacts of vineyard establishment, planting, and disposal phases. Moreover, to accurately assess efficiency using DEA, the vineyards must be healthy and in full production to capture their peak performance during their most productive years.

The system boundaries are illustrated in Figure 2, encompassing all relevant phases and associated inputs from grape production for each management system. A functional unit of 1 hectare of agricultural land is chosen to standardize comparisons across grapevine management systems, as this unit is commonly used in such studies to reduce variations [42,43]. In addition to assessing environmental impacts, production cost data will be evaluated by integrating shadow prices for the environmental impacts and incorporating them into the overall results. The efficiency analysis will determine whether these additional costs influence the final efficiency scores and highlight the significance of their impact. The study focuses on the three main viticultural management systems of the area, integrating five organic, four integrated, and five conventional vineyards.

2.3. Comprehensive Inventory Analysis

Accurate inventory data for various inputs, such as raw materials, energy consumption, and emissions, have driven the creation of credible databases, particularly for agricultural production. The inventory analysis regarding the primary energy usage, greenhouse gas (GHG) emissions, and relevant cost data is illustrated in

Table 1. Inventory analysis.

Inputs	Unit	Energy Content	Unit	GHGs	Unit	Cost	Remarks
Ν	MJ/kg	48.99	gCO2eq/kg	4524.41	EUR/kg	0.44–3.5	[44]
Ρ	MJ/kg	15.23	gCO2eq/kg	541.67	EUR/kg	0.49–2.5	[44]
Κ	MJ/kg	9.68	gCO2eq/kg	416.67	EUR/kg	0.33–3.5	[44]
Pesticides	MJ/kg	268.4	gCO2eq/kg	12,003.33	EUR/kg	10	[44]
Fungicides	MJ/kg	99.00	gCO2eq/kg	3900.00	EUR/kg	3.86–117.4	[45,46]
Herbicides	MJ/kg	418.00	gCO2eq/kg	9100.00	EUR/kg	3.47–8	[45,47]
Lubricants	MJ/kg	53.28	gCO2eq/kg	947.00	EUR/L	6	[44]
Diesel	MJ/kg	56.80	gCO2eq/MJ	95.10	EUR/L	1.602	[44,48,49]
Petrol	MJ/kg	60.20	gCO2eq/MJ	93.30	EUR/L	1.852	[44,48,49]
Electricity	MJ/MJ	2.73	gCO2eq/MJ	243.49	EUR/kWh	0.1941	[44,50]
Tractor	MJ/h	16.42	gCO2eq/h	9800	EUR/h	5.22–11.65	[51,52]
Human	MJ/h	1.80	gCO2eq	-	EUR/h	4.5–7.3	[51]
Machinery	MJ/h	0.10–35.05	gCO2eq/h	0.10–190	EUR/h	0.32–4.45	[53,54,55]
Irrigation system	MJ/ha	373.7	gCO2eq	-	EUR/h	0.05	[56]
Use of diesel	MJ	-	gCO2eq/MJ	0.9	EUR	-	[44]
Supplies	MJ/t·km	0.87	gCO2eq/t·km	71	EUR	-	[44,57]
Biomass	MJ/t·km	0.81	gCO2eq/t·km	71	EUR	-	[44,57]

. The analysis is concentrating on the cultivation phase of grapes, neglecting emissions from land use change, since the primary objective of the study is to evaluate the core production phase across the three management systems. The Sixth Assessment Report of the IPCC has updated the radiative efficiencies and metrics used for calculating global warming potential (GWP₁₀₀) in CO₂ equivalents: CO₂ = 1, CH₄ = 27.9, and N₂O = 273. A 100-year time horizon is adopted, considering both short- and mid-term effects, while indirect emissions from nitrogen fertilization and other factors like lubricant use and transportation are also estimated based on similar studies of viticultural projects [42,43].

2.4. Economic and Environmental Assessment

The development of a single unified indicator that converts climate pollutants into CO₂ eq. is determined by two key elements: the conversion factor and the corresponding quantity of each pollutant [58]. Global warming potential (GWP) is the conversion factor for a specified time horizon (100 years) and EF is referring to the gases emitted of each pollutant j related to each management system i as outlined below:

$${ENI}_i=\sum _{n=1}^i{EF}_{i,j}\times {GWP}_{100}$$

(1)

ENI_i is the environmental impacts in CO₂ equivalents for each management system i. Using an additive aggregation framework, the emissions from all relevant inputs are accounted for through an emissions factor approach [59]. Agricultural production encompasses various on-field activities, with associated emissions reflecting the inputs detailed in Table 1. Apart from the environmental aspects, annual estimates that account for cash flow variations, investment costs, and revenue throughout the vineyard lifecycle can be accurately calculated using a life cycle costing (LCC) approach [43,60,61,62]. The cost analysis of vineyards under different management systems emphasizes differences in field activities and inputs, with a discount rate set at 4% to reflect market conditions. In this context, the capital service cost of machinery equipment is also influenced, as it includes the equivalent costs of maintenance, insurance, and purchase expenses over the equipment’s lifetime. In agriculture, this approach involves breaking down a project into distinct activities and assigning the corresponding production input costs to each activity, based on the actual consumption levels associated with each operation as follows.

$${ECI}_i=\sum C_{rm}+\sum C_{la}+\sum C_{mc}+\sum C_l+\sum C_e+\sum C_{ov}+\sum C_{sp}$$

(2)

where ECI_i denotes the economic impact from the farmer standpoint for each management system, C_rm represents the total cost of acquiring raw materials, C_la denotes labor expenses, C_mc refers to capital service costs associated with machinery usage (including maintenance, insurance, and depreciation), C_l reflects landownership costs, C_e captures the total energy costs (diesel, petrol, and electricity), C_ov accounts for overhead costs across each system, and C_sp represents the added environmental cost in the form of shadow prices. The shadow prices for global warming, expressed in EUR per kilogram of CO₂ equivalents, are derived from De Bruyn et al. [33] and have been adjusted to 2024 values using inflation rates from the CSO’s PxStat Open Data Platform [63], resulting in a price of 0.492 EUR/kg CO₂ eq. Given the ongoing discussions surrounding the Polluter Pays Principle (PPP) in EU environmental policies and initiatives, the associated environmental burdens are being increasingly shifted onto the polluters, specifically farmers for this study [64]. The cost data for the selected vineyards have been meticulously documented and entered into the Activity-Based Costing (ABC^©) software v.2.1.2.0 [65], a tool designed for analyzing investment project costs in agriculture.

2.5. Data Envelopment Analysis

The evaluation of efficiency for services, institutions, public entities, and of course farms was a substantial challenge in previous decades. In 1978, Charnes et al. [66] presented a solution to this problem through the development of DEA. The importance of DEA is based on its ability to comparatively assess the efficiency or performance of individual decision-making units (DMUs) within a target group of interest that operates within specific application domains, such as agriculture [67]. This provides a valuable tool for evaluating the relative performance of individual DMUs within a particular industry. The technical efficiency of a DMU can be calculated as the ratio of the weighted sum of outputs to the weighted sum of inputs, as expressed by the following general equation [68]:

$${\displaystyle \frac{{\sum }_{n=1}^s{w}_n{y}_n}{{\sum }_{i=1}^m{v}_i{x}_i}}$$

(3)

where y_n is the amount of output n, w_n is the respective weight, x_i is the amount of input I, v_i is the weight of the respective input i. The challenge of determining a weight set when computing the efficiency score can be addressed through the application of linear programming techniques. More specifically, the optimization of the Equation (3) is achieved by maximizing its value subject to the constraints imposed by the efficiency scores of other DMUs (j = 1, …, m) [69] as follows:

$$max{\displaystyle \frac{{\sum }_{n=1}^s{w}_n{y}_{nj0}}{{\sum }_{i=1}^m{v}_i{x}_{i0}}}={\theta }_0$$

$$s.t {\displaystyle \frac{{\sum }_{n=1}^s{w}_n{y}_{nj}}{{\sum }_{i=1}^m{v}_i{x}_{ij}}}\le 1, \forall j$$

(4)

$$w,v\ge 0, \forall n and \forall i$$

Nevertheless, agricultural production depends scarce resources, making input-oriented DEA models more suitable for optimizing resource use by minimizing inputs in the production phase [70]. An input-oriented DEA model is expressed in the form of a linear programming formulation as follows [71]:

$$\mathit{max}\sum _{n=1}^s{w}_n{y}_{n0}={\theta }_0^{CRS}$$

$$s.t \sum _{n=1}^s{v}_i{x}_{i0} =1$$

(5)

$$\sum _{n=1}^s{w}_n{y}_{nj}-\sum _{i=1}^m{v}_i{x}_{ij}\le 0, \forall j$$

$$w,v\ge 0, \forall n and \forall i$$

The above model, Charnes, Cooper, and Rhodes (CCR model), operates under the assumption of Constant Returns to Scale (CRS), indicating that outputs increase proportionally with the corresponding inputs [66], while the Banker, Charnes, and Cooper (BCC) model is based on the assumption of variable returns to scale (VRS), allowing for the possibility of increasing, constant, or decreasing returns to scale [72]. To incorporate variable returns to scale, an additional constant variable (u₀) is introduced as follows:

$$\mathit{max}\sum _{n=1}^s{w}_n{y}_{n0}-u_0={\theta }_0^{VRS}$$

$$s.t \sum _{n=1}^s{v}_i{x}_{i0} =1$$

(6)

$$\sum _{n=1}^s{w}_n{y}_{nj}-u_0-\sum _{i=1}^m{v}_i{x}_{ij}\le 0, \forall j$$

$$w,v\ge 0, \forall n and \forall i$$

These techniques allow for the calculation of relative efficiency scores of DMUs and the identification of optimal practices for enhancing efficiency. A score of 1 signifies that the DMU is functioning at the efficiency frontier, while a score less than 1 indicates suboptimal performance, characterized by inefficiency. Finally, scale efficiency (SE) captures the impact of the farm’s size on its overall efficiency, indicating that a portion of inefficiency may be due to the size of the DMU [73] as follows:

$${SE}_j={\displaystyle \frac{{\theta }_0^{CRS}}{{\theta }_0^{VRS}}}$$

(7)

When SE_j = 1, the DMU_j operates at its optimal scale, being efficient under both CRS and VRS models. Conversely, when SE_j < 1, the DMU_j is not at its optimal scale, indicating potential efficiency improvements by either increasing or decreasing its operational scale. The technical efficiency scores from the non-increasing returns to scale (NIRS) and VRS models could be compared to identify increasing and non-increasing returns to scale. A difference in these scores indicates that the DMU operates under increasing returns to scale, while identical scores suggest the presence of decreasing returns to scale [74]. The data analysis for the DEA implementation, as detailed in the Results section, was performed using RStudio 2023.12.1 + 402 [75].

The inputs of the studied system consist of three fundamental production factors: land measured in hectares, labor quantified in euros per hectare, and capital represented by costs associated with the different management practices. Gross profit was selected as the output variable. This analysis was subsequently repeated with the inclusion of an additional input—shadow prices, which account for environmental impacts derived from the LC analysis, thereby introducing an extra cost element.

3. Results

3.1. Resource Inputs and Yield of the Selected Management Systems

In this study, the calculation of emissions, energy consumption, and the formulation of the DEA for each management system were conducted using inventory data and information provided by local vineyard owners. The analysis of viticulture management systems revealed notable differences in both economic and environmental performance metrics, as presented in

Table 2. Management scheme of viticulture management systems.

Type	Land (ha)	Yield (t/ha)	Human Labour (h/ha)	Residues (t/ha)	Diesel (L/ha)
O1	7.20	5.05	215	1.5	165.00
O2	7.50	4.65	230	1.4	172.00
O3	6.00	6.11	225	1.7	185.00
O4	5.00	6.17	250	1.7	210.00
O5	2.40	5.33	425	1.5	142.00
I1	2.50	5.73	200	1.4	130.00
I2	3.00	5.23	208	1.4	145.00
I3	2.90	8.53	450	3.9	255.00
I4	1.90	8.02	447	4.2	255.00
C1	11.00	9.07	240	3.2	165.00
C2	9.50	10.1	270	3.5	180.00
C3	8.00	10.3	240	3.0	165.00
C4	7.00	9.90	270	3.2	180.00
C5	5.90	8.48	372	2.9	250.00

. Yield varied significantly among the three management systems, with Conventional (C) practices consistently producing higher outputs as expected. The highest yield was observed for C2 and C3 vineyards at 10.1 t/ha and 10.3 t/ha, respectively. Organic (O) management and specifically vineyard O2 produced the lowest yield (4.65 t/ha), while Integrated (I) practices generally yielded intermediate results. In terms of resource inputs, O vineyards were characterized by higher labor demands, especially in vineyard O5, which required 425 h/ha, whereas I management showed significant variation, with I3 and I4 requiring labor inputs on par with O systems (450 h/ha and 447 h/ha, respectively). Diesel consumption followed a similar pattern, with the highest usage observed in I3, I4, and C5 systems, reaching up to 255 L/ha.

3.2. Energy Demand and GHG Emissions

Agricultural operations, such as spraying, pruning, harvesting etc., are associated with a range of energy consumption and input management, thus generating respective environmental impacts. The magnitude of these operations alters the total GHG emissions, based in the implemented management system. Energy consumption in kWh/ha and GHG emissions in kg CO₂ eq./ha for each management system are graphically presented in Figure 3 through a dual Y-axis bar-line plot.

The O management systems present relatively lower energy consumption, ranging from 4545.98 kWh/ha (O5) to 6572.69 kWh/ha (O4), while corresponding emissions are also lower, between 1358.42 (O1) and 1795.35 kg CO₂ eq./ha (O4). On the contrary, the I and C systems show significantly higher energy demand, and the relevant emissions are higher as well. More specifically, energy consumption for the I systems varies from 9156.51 kWh/ha (I3) to 9841.15 kWh/ha (I2), while emissions range from 2961.33 (I3) to 3647.45 kg CO₂ eq./ha (I2). Among the C systems, C5 stands out with the highest energy consumption (12,108.91 kWh/ha) and emissions (3394.90 kg CO₂ eq./ha).

3.3. Economic Analysis and Shadow Prices

The economic analysis of each management system was evaluated via an LCC framework, accounting for all relevant production costs. Land costs encompass expenses related to acquiring agricultural areas for cultivation, while energy costs are associated with the consumption of diesel, petrol, and electricity. Additionally, labor costs refer exclusively to human labor, and machinery costs represent the capital service expenses for all equipment utilized in viticultural practices (e.g., tractors, vine trimmers, fertilizer distributors, etc.). Raw material costs include all inputs related to management practices, such as fertilizers, pesticides, herbicides, and similar materials (

Table 3. Cost analysis of several inputs for each management system.

Type	Unit	Land	Raw Materials	Energy	Labor	Machinery
O1	EUR/ha	300.00	154.00	343.74	1245.05	257.51
O2	EUR/ha	300.00	196.85	368.93	1312.55	257.51
O3	EUR/ha	300.00	168.60	423.53	1356.35	296.62
O4	EUR/ha	300.00	252.45	474.28	1503.55	330.80
O5	EUR/ha	150.00	136.08	346.21	2335.97	254.85
I1	EUR/ha	200.00	420.80	300.72	848.25	335.41
I2	EUR/ha	200.00	465.70	326.36	910.25	362.78
I3	EUR/ha	200.00	525.00	764.99	1264.25	493.65
I4	EUR/ha	200.00	175.00	735.88	1572.00	455.30
C1	EUR/ha	200.00	418.00	578.87	1378.00	626.08
C2	EUR/ha	200.00	485.00	384.74	1267.25	453.05
C3	EUR/ha	200.00	431.50	368.72	1177.25	453.05
C4	EUR/ha	200.00	485.00	433.47	1410.25	504.74
C5	EUR/ha	200.00	500.36	604.86	1878.00	455.30

).

The current economic analysis highlights significant variations in costs across land, raw materials, energy, labor, and machinery. Among the O systems, labor costs are particularly prominent. More specifically, O5 shows the highest labor cost at EUR 2335.97/ha, substantially exceeding the other O vineyards, which range from EUR 1245.05/ha (O1) to EUR 1503.55/ha (O4). For the C systems, there is a noticeable increase in costs as management intensity escalates. C5 exhibits the highest total expenditure for multiple categories, particularly in labor (EUR 1878.00/ha) and energy (EUR 604.86/ha), thus focusing on high-input and high-output production strategy. Additionally, raw material costs peak in C5 at EUR 500.36/ha, emphasizing reliance on external inputs such as fertilizers and pesticides. The I systems demonstrate varying resource efficiencies, with I1 and I2 generally showing lower energy and labor costs in comparison to O systems.

Conversely, C systems show elevated costs in most categories, while I systems present a balanced distribution of costs, with raw materials and machinery representing significant expenditures. I3 demonstrates the highest energy cost at EUR 764.99/ha, indicating a more resource-intensive approach compared to the other I scenarios. Additionally, I4 shows the highest labor cost at EUR 1572.00/ha, reflecting increased manual intervention during cultivation. In addition to the associated costs,

Table 4. Economic gains and shadow prices for each management system.

Type	Unit	Profit	Sales	Shadow Prices
O1	EUR/ha	2143.79	3289.31	668.34
O2	EUR/ha	1741.33	3025.12	721.93
O3	EUR/ha	2623.89	4016.94	770.26
O4	EUR/ha	2615.21	4348.72	883.31
O5	EUR/ha	2235.68	4268.79	763.40
I1	EUR/ha	2362.92	4016.12	1726.38
I2	EUR/ha	2376.74	4189.85	1794.55
I3	EUR/ha	2751.78	5979.67	1456.97
I4	EUR/ha	2902.94	6021.12	1600.81
C1	EUR/ha	3123.04	5899.21	1699.77
C2	EUR/ha	3213.39	5551.45	1801.32
C3	EUR/ha	2968.82	5147.36	1727.09
C4	EUR/ha	3354.55	5936.03	1797.54
C5	EUR/ha	3857.95	8476.47	1670.29

also presents the economic gains, alongside the environmental impacts transmuted into shadow prices.

Profit values vary significantly among the systems studied. The C systems outperform both O and I in terms of profitability, with C5 reaching the highest profit of EUR 3857.95/ha, highlighting the economic advantages of more intensive production methods. O systems achieve sales between EUR 3025.12/ha (O2) and EUR 4348.72/ha (O4), reflecting a balance between lower yields and premium pricing, while I systems show higher sales (e.g., I4 at EUR 6021.12/ha) driven by a combination of increased yield and sustainable practices.

When considering shadow prices—which represent the environmental impacts converted into economic terms—the results reveal that O systems have the lowest shadow prices, ranging from EUR 668.34/ha (O1) to EUR 883.31/ha (O4). Nevertheless, I and C systems illustrate higher shadow prices, particularly in I2 (EUR 1794.55/ha) and C4 (EUR 1801.32/ha), indicating a trade-off between increased productivity and environmental impact. This concept is further elucidated in Figure 4, which displays a comparative analysis of total costs both with and without the inclusion of shadow prices.

Beginning with the C management systems, the results indicate that these management systems generally incur higher added costs, generating deviations peaking at 64.00% (C3) and 63.03% (C2), highlighting the variability and higher uncertainty associated with intensive practices. Similarly, the added costs that stand out for I systems, e.g., I1 and I2 where the total added costs reach EUR 3899.57 and EUR 4127.66/ha respectively, experiencing considerable cost fluctuations, likely due to variability in input usage. Despite having the lowest overall costs, O systems still present notable deviations. For example, O5 has a deviation of 23.25%, the lowest across all scenarios, reflecting greater stability and predictability in sustainable production phases. O systems, while more stable, involve relatively lower shadow prices, highlighting their potential as a more sustainable and economically viable approach under certain conditions.

3.4. Efficiency Assessment

DEA was used in order to evaluate and compare performance scores among the vineyard management systems. The outcomes of the DEA, applying both the input-oriented CCR and BCC models, are presented in

Table 5. Efficiency indices of farmers for each management system (Increasing Returns to Scale—IRS, Decreasing Returns to Scale—DRS and Scale Efficient—SE).

Type	CRS Model	VRS Model	NIRS Model	Scale Efficiency	Returns to Scale
O1	0.82	0.83	0.82	0.98	IRS
O2	1.00	1.00	1.00	1.00	SE
O3	0.93	0.94	0.93	0.98	IRS
O4	0.85	0.88	0.85	0.97	IRS
O5	0.93	1.00	0.93	0.93	IRS
I1	1.00	1.00	1.00	1.00	SE
I2	0.95	0.96	0.96	1.00	SE
I3	0.86	0.96	0.86	0.90	IRS
I4	0.84	1.00	0.84	0.84	IRS
C1	0.94	1.00	1.00	0.94	DRS
C2	1.00	1.00	1.00	1.00	SE
C3	0.98	0.99	0.99	0.99	DRS
C4	1.00	1.00	1.00	1.00	SE
C5	1.00	1.00	1.00	1.00	SE

for all management systems under study. The efficiency analysis reveals a diverse range of performance outcomes, with O2, I1, I2, C2, C4, and C5 achieving perfect efficiency scores under both models and indicating optimal performance at their current scale. In contrast, most of the O management systems demonstrated efficiencies below 1.00, suggesting potential benefits from scaling up their operations.

However, these systems also showed high scale efficiency scores, including O1, O3, O4, O5, I3, and I4, suggesting they could gain efficiency from expanding their scale through Increasing Return to Scale (IRS). Conversely, C1 and C3 were classified experiencing Decreasing Returns to Scale (DRS), indicating that these systems are approaching their operational limits. The technical efficiency among the different management systems shows minimal variation, reflecting a high level of efficiency among grape producers in the studied area. In addition to the previously mentioned analysis, the input-oriented CCR and BCC models were rerun with the inclusion of an additional input variable, which represents the environmental burden converted into costs using shadow prices. The results of this updated analysis are presented in

Table 6. Efficiency indices of farmers for each management system integrating shadow prices (Increasing returns to scale—IRS, decreasing returns to scale—DRS, and scale efficient—SE).

Type	CRS Model	VRS Model	NIRS Model	Scale Efficiency	Returns to Scale
O1	0.92	0.92	0.92	1.00	SE
O2	1.00	1.00	1.00	1.00	SE
O3	1.00	1.00	1.00	1.00	SE
O4	0.92	0.95	0.92	0.96	IRS
O5	0.96	1.00	0.96	0.96	IRS
I1	0.55	1.00	0.55	0.55	IRS
I2	0.47	0.82	0.47	0.57	IRS
I3	0.75	0.99	0.75	0.76	IRS
I4	0.66	1.00	0.66	0.66	IRS
C1	0.76	1.00	1.00	0.76	DRS
C2	0.82	0.97	0.97	0.84	DRS
C3	0.77	0.78	0.78	0.98	DRS
C4	0.82	0.83	0.82	0.99	IRS
C5	1.00	1.00	1.00	1.00	SE

.

The integration of shadow prices into the DEA models reveals that most C vineyard management systems do not maintain high efficiency when accounting for environmental costs. The impact on their efficiency is significant for both models, except from C5 which remains SE. Nevertheless, several O systems—namely O1, O2, and O3—exhibit perfect scale efficiency (1.00), indicating that they operate optimally when environmental impacts are factored in. Simultaneously, systems O4, O5, I3, and I4 exhibit IRS, indicating that these systems have not reached optimal scale efficiency again.

These findings underscore the impact of environmental costs on relative efficiency among viticultural management systems and highlight their positive impact on O management systems and negative on C and I systems. If the environmental burden is entirely attributed to the farmers, inefficiencies will significantly affect the average scale efficiency of the C and I management systems. The extensive use of agrochemicals, which is the primary distinguishing factor between the studied systems, has a substantial impact on the efficiency of farms when translated into an economic burden.

4. Discussion

This study primarily focuses on identifying the environmental and socio-economic differences among the three main viticultural management systems in the Drama region of Northern Greece. The findings of this study align with previous researches, emphasizing the lower environmental impacts of organic viticulture compared to conventional management systems [15,17,47]. A similar pattern is observed in yield results, where conventional and integrated viticultural management systems demonstrate higher yields compared to organic systems [13,18]. Nevertheless, as noted by Ghiglieno et al. [19], significant differences in fuel consumption were observed between organic and conventional viticultural management systems in Northern Italy, while Rouault et al. reported that organic systems have greater environmental impacts compared to integrated systems [76]. In contrast, the current study found average fuel consumption values to be relatively consistent among the investigated systems, while emissions were significantly higher in integrated and conventional systems.

Furthermore, the economic analysis revealed lower costs for organic viticultural management systems compared to integrated and conventional systems. As anticipated, agrochemical-related expenses were reduced in organic management. However, unlike findings from other studies, labor costs in this research did not show significant differences across the management systems [77,78]. This could be attributed to the use of newer organic-approved substances and pest management practices, which are more effective and reliable compared to methods used in previous years [79]. In addition to the points mentioned above, the current study also assesses added costs in the form of shadow prices, incorporating the PPP in viticulture. The PPP in agriculture is not a novel concept [80], yet there are various challenges regarding the pricing of these emissions and the allocation of costs among different stakeholders [81]. Nevertheless, EU policies face significant challenges in reducing GHG emissions in the agriculture sector, with the implementation of the PPP in this area remaining inadequate [82]. Measures that rely exclusively on the PPP could potentially impose substantial negative impacts on the broader economy [83]. Thus, the focus should shift from relying solely on the PPP to adopting provider gets principle measures [84], which offer incentives to facilitate a gradual transition toward sustainability.

To address hidden costs in agri-food systems related to environmental and social impacts, a range of methods has been demonstrated [85]. In this context, shadow prices have been employed to evaluate the emission reduction cost [86] and the economic valuation of greenhouse gases from agricultural activities [29,87]. The findings suggest that the added costs from shadow prices could represent a substantial financial burden for farmers, primarily influenced by their input usage and on-field agricultural operations. This economic impact not only affects the overall cost for each management system but also influences the efficiency of each system. A dual sustainability objective for economic and environmental aspects of viticultural systems should prioritize optimizing key inputs that have a substantial influence on system performance. Notably, variable rate drip irrigation and fertigation [88] and the integration of vineyards and livestock systems [89] have demonstrated significant impacts in this regard.

Although small data samples in DEA are criticized for uncertainty and upward bias [90,91], the focus of this study is on the robust comparison of the investigated viticultural management systems rather than exclusively on vineyard efficiency. Furthermore, efficiency estimates for the three agricultural production factors—labor, capital, and land—can fluctuate their levels and rankings when evaluated using different metrics [92,93]. Consequently, this study emphasizes the comparison of similar vineyards within the same region, analyzing their DEA scores under uniform conditions to better understand the potential impacts of environmental burdens if these are entirely borne by the farmers. The conversion of environmental impacts into economic costs reveals substantial implications, underscoring the need for active engagement from various stakeholders to maintain and advance sustainability efforts. This supports the concept of a self-aware society that acknowledges the advantages of more sustainable viticultural systems and is willing to pay more for good agricultural and environmental conditions [94]. A collaborative effort among stakeholders is crucial to support and incentivize the adoption of sustainable viticultural practices, highlighting a significant connection between environmental and economic relationships within viticulture.

5. Conclusions

This study examined the environmental and economic performance of three viticultural management systems employed by farmers in the Drama region of Northern Greece. The analysis revealed that organic systems have a lower environmental impact, although they yield lower income and exhibit lower efficiency compared to conventional systems. However, the integration of shadow prices into the economic analysis significantly altered the efficiency of organic viticultural management systems, highlighting a shift from conventional practices towards more sustainable agricultural methods. This transition should be gradual, with policy measures accompanying each phase to ensure a smooth and effective shift. Therefore, attention should move beyond exclusively relying on the polluter pays principle (PPP) and instead embrace the provider gets principle measures. This creates a framework that encourages farmers to invest in sustainable practices, facilitating a gradual transition towards sustainability and ensuring that all parties benefit from environmentally responsible practices.

Future research should incorporate a larger dataset to provide a clearer perspective on efficiency. Additionally, integrating additional inputs into the DEA analysis, such as non-renewable energy, could elucidate the advantages and disadvantages of mechanical labor. This approach has the potential to significantly influence the assessment of energy use and environmental performance in vineyards. Finally, investigating super-efficiency could further refine the discriminative power of DEA, providing a more nuanced and detailed ranking of the management systems.